Defense of the doctoral thesis in Artificial Intelligence by Ms. Yasmine GHAZLANE

Euromed University of Fes (UEMF) is pleased to inform the public of the defense of the doctoral thesis in ‘‘ Artificial Intelligence ’’

The thesis defense will take place on Wednesday May 29, 2024 at 4:00 PM at UEMF

Location : Amphitheater 4 in Building 4

The thesis will be presented by Ms. Yasmine GHAZLANE Under the theme:

“A smart anti-drone architecture based on a multi-agent system with the development of airborne targets detection and identification models using advanced artificial intelligence and computer vision techniques” 
 

Abstract:

This doctoral thesis presents an objective investigation into the realm of anti-drone technology, addressing critical gaps in existing defense mechanisms against the escalating proliferation of unmanned aerial vehicles, commonly known as drones. An extensive and meticulous literature review is conducted as a first step toward examining the evolving landscape of anti-drone systems, identifying deficiencies and opportunities for transformation. In this study, artificial intelligence techniques and cutting-edge technologies are employed to propose and elucidate a novel and suitable anti-drone architecture. Leveraging the robustness of the Blockchain approach, the adaptability of multi-agent systems, and the precision of expert system, this thesis outlines a sophisticated framework poised to revolutionize the landscape of drone defense mechanisms.  The architecture embodies a paradigm shift, fostering agility and resilience in identifying the most encountered airborne targets in real-time and countering harmful targets across multifaceted operational domains through the use of an AI-based electromagnetic neutralization technique. Furthermore, this thesis unveils the development of highly efficient real-time detection and identification models, designed to seamlessly integrate within the proposed architecture. Using advanced algorithms and deep learning techniques, the developed models identify the most encountered airborne targets promptly and accurately while satisfying the speed and performance compromise. In the existing literature, most of the attention has been centered on recognizing drones as unique airborne targets whereas the real challenge is to distinguish between drones and non-drone targets. To address this issue, we have developed an identification friend or foe backbone model able to classify the aerial targets in foe or friend categories by determining whether the aerial target is a drone or bird, respectively. To meet the antidrone requirements, artificial intelligence and computer vision approaches have been combined through transfer learning, data augmentation and other techniques in our model. Another contribution of this thesis is the study of the impact of depth on classification performance, which is demonstrated through our experiments. The identification friend or foe model is integrated as a backbone model within the system with the real-time detection module to enhance further the efficiency, accuracy, adaptability, and robustness of the overall anti-drone system to enable it to effectively identify and respond to potential threats in dynamic and complex environments and neutralize effectively the target.
 

This thesis will be presented before the members of the jury: